Color is a fundamental aspect of vision. Color is not in light but rather in the neural representation of light in eye and brain. This neural representation, and therefore the perceived color of a light, depends on the context of other lights also in view. Classically, context has been studied with uniform backgrounds. The proposed research investigates neural processes of color perception using context that has spatial structure. Spatial structure of context is a basic property affecting color appearance. For example, recent results from my laboratory show that the color of a light can be shifted substantially more by introducing a background pattern composed of two chromaticities than by a uniform background at either of the two chromaticities in the pattern. A fundamental unsolved problem, which is addressed by this research, is the level(s) of neural representation where context with spatial structure affects color perception. This is a central issue for understanding perception of natural scenes, in which most objects are seen within a complex patchwork of other light. In general, context can induce two qualitatively distinct types of shift in perceived color, either away from the appearance of nearby context (color contrast) or toward it (color assimilation). Initially, contextual patterns will be selected to study separately the neural processes that mediate contrast and that mediate assimilation. Later studies will integrate these results to achieve a broader understanding of color perception with more general types of context with spatial structure. Neural processes that mediate color vision will be inferred by measuring changes in the perceived color of a light caused by introducing strategically designed surrounding patterns. The experiments are devised to distinguish among competing hypotheses of neural representation (for example, receptor signals, postreceptoral chromatically-opponent neural pathways, or more chromatically selective, higher-order central mechanisms). Reduced color vision in patients typically is assessed with tests that use simple visual fields. Neural mechanisms affected only by complex spatial structure, which occurs in natural viewing, are not evaluated. A long-term goal of this research is to understand the many neural processes mediating color perception of complex visual stimuli, and to develop methods to assess the functioning of these neural mechanisms.